update model card README.md
Browse files
README.md
CHANGED
|
@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the natural_questions dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
-
- Loss: 0.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
@@ -41,22 +41,52 @@ The following hyperparameters were used during training:
|
|
| 41 |
- seed: 42
|
| 42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 43 |
- lr_scheduler_type: linear
|
| 44 |
-
- num_epochs:
|
| 45 |
|
| 46 |
### Training results
|
| 47 |
|
| 48 |
-
| Training Loss | Epoch | Step
|
| 49 |
-
|
| 50 |
-
| 2.0532 | 1.0 | 5104
|
| 51 |
-
| 1.8912 | 2.0 | 10208
|
| 52 |
-
| 1.7854 | 3.0 | 15312
|
| 53 |
-
| 1.6856 | 4.0 | 20416
|
| 54 |
-
| 1.5918 | 5.0 | 25520
|
| 55 |
-
| 1.5067 | 6.0 | 30624
|
| 56 |
-
| 1.4323 | 7.0 | 35728
|
| 57 |
-
| 1.365 | 8.0 | 40832
|
| 58 |
-
| 1.3162 | 9.0 | 45936
|
| 59 |
-
| 1.281 | 10.0 | 51040
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 60 |
|
| 61 |
|
| 62 |
### Framework versions
|
|
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the natural_questions dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.6267
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
|
|
| 41 |
- seed: 42
|
| 42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 43 |
- lr_scheduler_type: linear
|
| 44 |
+
- num_epochs: 40
|
| 45 |
|
| 46 |
### Training results
|
| 47 |
|
| 48 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 49 |
+
|:-------------:|:-----:|:------:|:---------------:|
|
| 50 |
+
| 2.0532 | 1.0 | 5104 | 0.2393 |
|
| 51 |
+
| 1.8912 | 2.0 | 10208 | 0.2284 |
|
| 52 |
+
| 1.7854 | 3.0 | 15312 | 0.2357 |
|
| 53 |
+
| 1.6856 | 4.0 | 20416 | 0.2487 |
|
| 54 |
+
| 1.5918 | 5.0 | 25520 | 0.2743 |
|
| 55 |
+
| 1.5067 | 6.0 | 30624 | 0.2586 |
|
| 56 |
+
| 1.4323 | 7.0 | 35728 | 0.2763 |
|
| 57 |
+
| 1.365 | 8.0 | 40832 | 0.2753 |
|
| 58 |
+
| 1.3162 | 9.0 | 45936 | 0.3200 |
|
| 59 |
+
| 1.281 | 10.0 | 51040 | 0.3127 |
|
| 60 |
+
| 1.308 | 11.0 | 57104 | 0.2947 |
|
| 61 |
+
| 1.241 | 12.0 | 62208 | 0.2941 |
|
| 62 |
+
| 1.1391 | 13.0 | 67312 | 0.3103 |
|
| 63 |
+
| 1.0334 | 14.0 | 72416 | 0.3694 |
|
| 64 |
+
| 0.9538 | 15.0 | 77520 | 0.3658 |
|
| 65 |
+
| 0.8749 | 16.0 | 82624 | 0.4009 |
|
| 66 |
+
| 0.8154 | 17.0 | 87728 | 0.3672 |
|
| 67 |
+
| 0.7533 | 18.0 | 92832 | 0.3675 |
|
| 68 |
+
| 0.7079 | 19.0 | 97936 | 0.4611 |
|
| 69 |
+
| 0.6658 | 20.0 | 103040 | 0.4222 |
|
| 70 |
+
| 0.595 | 21.0 | 108144 | 0.4095 |
|
| 71 |
+
| 0.5765 | 22.0 | 113248 | 0.4400 |
|
| 72 |
+
| 0.5259 | 23.0 | 118352 | 0.5109 |
|
| 73 |
+
| 0.4804 | 24.0 | 123456 | 0.4711 |
|
| 74 |
+
| 0.4389 | 25.0 | 128560 | 0.5072 |
|
| 75 |
+
| 0.4034 | 26.0 | 133664 | 0.5363 |
|
| 76 |
+
| 0.374 | 27.0 | 138768 | 0.5460 |
|
| 77 |
+
| 0.3434 | 28.0 | 143872 | 0.5627 |
|
| 78 |
+
| 0.3181 | 29.0 | 148976 | 0.5657 |
|
| 79 |
+
| 0.2971 | 30.0 | 154080 | 0.5819 |
|
| 80 |
+
| 0.275 | 31.0 | 159184 | 0.5649 |
|
| 81 |
+
| 0.2564 | 32.0 | 164288 | 0.6087 |
|
| 82 |
+
| 0.2431 | 33.0 | 169392 | 0.6137 |
|
| 83 |
+
| 0.2289 | 34.0 | 174496 | 0.6123 |
|
| 84 |
+
| 0.2151 | 35.0 | 179600 | 0.5979 |
|
| 85 |
+
| 0.2041 | 36.0 | 184704 | 0.6196 |
|
| 86 |
+
| 0.1922 | 37.0 | 189808 | 0.6191 |
|
| 87 |
+
| 0.1852 | 38.0 | 194912 | 0.6313 |
|
| 88 |
+
| 0.1718 | 39.0 | 200016 | 0.6234 |
|
| 89 |
+
| 0.1718 | 39.81 | 204160 | 0.6267 |
|
| 90 |
|
| 91 |
|
| 92 |
### Framework versions
|